Seasonal variation of venous thromboembolism in the subtropical climate of São Paulo, Brazil

Abstracts

BACKGROUND: The triggering factors of venous thromboembolic disease have been increasingly clarified. External causes may influence its occurrence, and some climactic factors have stood out. Nothing is known about such interference in our latitude. OBJECTIVES: To determine whether there are seasonal variations in venous thromboembolism in a hospital-based population in São Paulo, Brazil, which has subtropical climate. METHODS: Medical records of patients admitted to Hospital da Beneficência Portuguesa de São Paulo with the diagnosis of deep venous thrombosis or pulmonary thromboembolism were reviewed from January 1996 to October 2003. Cases were grouped in trimesters (first trimester = January, February and March; second trimester = April, May and June; third trimester = July, August and September; fourth trimester = October, November and December). They were also grouped as to warm and cold months, according to mean temperature (warm months = October through April; cold months = May through September). RESULTS: A total of 955 cases of venous thromboembolism were found during the study period. The ANOVA test was used for statistical analysis, showing no significant difference in the occurrence of venous thromboembolism considering the four trimesters. Separate analysis of deep venous thrombosis and pulmonary embolism incidence showed no differences either. Comparing warm and cold months, there was an increased incidence of deep venous thrombosis during warm months (p < 0.05, Mann-Whitney test). CONCLUSION: Venous thromboembolism is not clearly related to climatic variations. The influence of climate and temperature on blood coagulability is poorly understood and needs to be further studied.

Venous thrombosis; pulmonary embolism; epidemiology; climatology

CONTEXTO: Os fatores desencadeantes da doença tromboembólica venosa vêm sendo cada vez melhor identificados. Causas externas podem influir na sua ocorrência, e algum destaque tem sido dado a fatores climáticos. Nada se sabe quanto a essa interferência em nossa latitude. OBJETIVOS: Analisar se há diferença na incidência do tromboembolismo venoso de acordo com as estações do ano, num hospital da cidade de São Paulo, Brasil, cujo clima é categorizado como subtropical. MÉTODOS: Foi realizado trabalho retrospectivo de levantamento de dados a partir de prontuários de pacientes cujo diagnóstico de internação ou óbito foi de trombose venosa profunda ou tromboembolismo pulmonar, no período de janeiro de 1996 a outubro de 2003, no Hospital da Beneficência Portuguesa de São Paulo. Para comparação e estudo, os casos foram agrupados em trimestres (primeiro trimestre = janeiro, fevereiro e março; segundo trimestre = abril, maio e junho; terceiro trimestre = julho, agosto e setembro; e quarto trimestre = outubro, novembro e dezembro) e conforme sua ocorrência nos chamados meses quentes e frios, de acordo com a média de temperatura mensal (meses quentes = outubro a abril; meses frios = maio a setembro). RESULTADOS: Foram encontrados 955 casos de tromboembolismo venoso no período analisado. Foi utilizado o teste ANOVA para análise, que não revelou diferença estatisticamente significativa na incidência do tromboembolismo venoso de acordo com os trimestres. Quando analisados separadamente, também não se evidenciou significância estatística em relação ao tromboembolismo pulmonar e à trombose venosa profunda. Quando comparados os meses quentes e frios, observou-se aumento da incidência de trombose venosa profunda nos meses quentes (p < 0,05, teste de Mann-Whitney). CONCLUSÃO: O tromboembolismo venoso é uma doença que não tem uma relação bem estabelecida com as variações climáticas. A influência da temperatura ambiental na coagulabilidade ainda precisa ser amplamente estudada.

Trombose venosa; embolia pulmonar; epidemiologia; climatologia

ORIGINAL ARTICLE

Seasonal variation of venous thromboembolism in the subtropical climate of São Paulo, Brazil

Daniela Kleinfelder^I; Jó Luis Andrade^II; Sascha Werner Schlaad^III; Francine Correa Carvalho^IV; Bonno van Bellen^V

^I

^IIGraduate Program in Integrated Vascular Surgery, BPSP, São Paulo, SP, Brazil. Professor, General Surgery, Faculdade de Medicina, Universidade Estadual do Sudoeste da Bahia (UESB), Vitória da Conquista, BA, Brazil

^IIIGraduate Program in Integrated Vascular Surgery, BPSP, São Paulo, SP, Brazil

^IVPhD in Cardiology, Instituto do Coração, Faculdade de Medicina da Universidade de São Paulo (InCOR-FMUSP), São Paulo, SP, Brazil. Cardiologist, Vascular Surgery and Angiology Service, BPSP, São Paulo, SP, Brazil

^VProfessor, Peripheral Vascular Diseases, Universidade Estadual de Campinas (UNICAMP), Campinas, SP, Brazil. Head, Vascular Surgery and Angiology Service, BPSP, São Paulo, SP, Brazil. Coordinator, Graduate Program in Integrated Vascular Surgery, BPSP, São Paulo, SP, Brazil

Correspondence

ABSTRACT

Background: The understanding of the triggering factors of venous thromboembolic disease has presented important improvements during the last years. External causes may influence its occurrence, and some climatic factors have also been mentioned to interfere with its occurrence. There are not studies about these specific interference in the subtropical climate.

Objectives: To determine whether there are seasonal variations in the incidence of venous thromboembolism in a hospital-based population in São Paulo, Brazil, which has subtropical climate.

Methods: Medical records of patients admitted to Hospital da Beneficência Portuguesa de São Paulo with the diagnosis of deep venous thrombosis or pulmonary thromboembolism were reviewed from January 1996 to October 2003. Cases were grouped in trimesters (first trimester = January, February and March; second trimester = April, May and June; third trimester = July, August and September; fourth trimester = October, November and December). They were also grouped as to warm and cold months, according to mean temperature (warm months = October through April; cold months = May through September).

Results: A total of 955 cases of venous thromboembolism were found during the study period. The ANOVA test was used for statistical analysis, showing no significant difference in the occurrence of venous thromboembolism considering the four trimesters. Separate analysis of deep venous thrombosis and pulmonary embolism incidence showed no differences either. Comparing warm and cold months, there was an increased incidence of deep venous thrombosis during warm months (p < 0.05, Mann-Whitney test).

Conclusion: Venous thromboembolism is not clearly related to climatic variations. The influence of climate and temperature on blood coagulability is poorly understood and needs to be further studied.

Keywords: Venous thrombosis, pulmonary embolism, epidemiology, climatology.

Introduction

Venous thromboembolism (VTE) is a major cause of morbidity and mortality in Western countries.¹ Only in the USA more than 100 cases/100,000 people/year are reported.² In Brazil and in other Latin American countries, there are few studies on VTE epidemiology.³ Maffei et al. estimated an incidence of 0.6 cases of deep venous thrombosis (DVT) for each 1,000 inhabitants/year in the municipality of Botucatu.⁴ Amari et al. found a 6% incidence of pulmonary thromboembolism (PTE) in a necropsy study at Santa Casa de São Paulo.⁵

Mean population age has increased, as well as the absolute number of VTE cases.⁶ Some of these cases will be manifested as PTE. In England, PTE is the cause of 200,000 deaths/year, accounting for 1/10 in-hospital deaths⁷ and is the most common avoidable cause of death in hospitalized patients in the USA.⁸ Of all PTE cases, 10% are fatal in less than 1 hour after event onset⁹ and approximately 5% of patients with PTE develop pulmonary hypertension.⁹ Chronic venous stasis syndrome affects 11% of patients with VTE, evolving to varicose ulcers in 3.7 ± 0.9% of patients in 20 years.⁹

Much has been done regarding the development of methods and drugs for VTE prophylaxis, and consequently there has been significant progress regarding understanding of causes and epidemiological phenomena of the disease.

Many studies have suggested variation of VTE incidence according to seasons of the year.^2,710-16 However, such studies were performed in countries with temperate climate. No study conducted in subtropical climate was found.

This study aims at analyzing whether there is a difference in VTE incidence according to seasons of the year at a hospital in São Paulo, located just on the Tropic of Capricorn, which is the limit between tropical and temperate climate.

Population and methods

A retrospective survey of medical records of patients whose diagnosis of hospitalization or death was VTE or PTE was performed from January 1996 to October 2003 at Hospital da Beneficência Portuguesa de São Paulo.

For comparison and analysis, cases were grouped according to trimesters:

- First trimester: January, February and March;

- Second trimester: April, May and June;

- Third trimester: July, August and September;

- Fourth trimester: October, November and December.

The ANOVA test was used in the analysis.

The year was also divided in warm and cold months. Such division was performed considering the mean monthly temperature ≥ 20 ºC (warm months) and < 20 ºC (cold months), respectively. Warm months are between October and April, and cold months are between May and September. Data were obtained from information on the website The Weather Channel¹⁷. The Mann-Whitney test was used for statistical analysis.

Statistical significance was set in p < 0.05.

Results

There were 955 cases of VTE: 453 cases of DVT and 502 cases of PTE (mean of 10.2 cases of VTE per month, standard deviation of 4.9).

In the first trimester, there were 253 cases of VTE (26.5%): 143 cases of DVT (31.6%) and 110 cases of PTE (21.9%). In the second trimester, there were 232 cases of VTE (24.3%): 100 cases of DVT (22.1%) and 132 cases of PTE (26.3%). In the third trimester, there were 223 cases of VTE (23.3%): 94 cases of DVT (20.7%) and 129 cases of PTE (25.7%). In the fourth trimester, there were 247 cases of VTE (25.9%): 116 cases of DVT (25.6%) and 131 cases of PTE (26.1%).

The ANOVA test did not show statistically significant difference in VTE incidence according to the trimesters. When analyzed individually, there was no statistical significance in relation to PTE and DVT.

In warm months (October-April, mean temperature > 20 ºC) there were 578 cases of VTE (60.5%): 283 cases of PTE (56.4%) and 295 cases of DVT (65.1%). In cold months (May-September, mean temperature < 20 ºC) there were 377 cases of VTE (39.5%): 219 cases of PTE (43.6%) and 158 cases of DVT (34.8%). The Mann-Whitney test showed increase in DVT cases in warm months (p < 0.05) (Figure 1).

Discussion

Many studies have reported higher incidence of death and hospitalizations due to coronary disease and stroke during the winter.^10,18 Similarly, some studies have shown higher incidence of fatal PTE in the winter than in the summer,^2,7,11,18 well documented in a meta-analysis of 23 studies involving 11,000 cases.^12,13 In an English necropsy study, there was a higher incidence of PTE in the fall and spring.¹⁴

Bounameaux et al.¹⁰ in 1996 found no seasonal variation in DVT incidence in a retrospective study of 9,208 patients from 1989 to 1994 in Geneva, a city with large temperature variations between the summer and the winter, as well as the French study by Galle et al., performed in 1998,¹⁵ and the American study by Stein et al.¹⁶ The latter evaluated seasonal variations of VTE over a 21-year period in regions with wide temperature variations and in areas with little climatic difference.

However, Boulay et al.¹³ in 2001 published a study conducted in France that included 62,237 patients, showing the existence of a seasonal variation in VTE occurrence, with number of hospitalizations 18 and 22% higher than the mean in cold months for DVT and PTE, respectively.

The likely reasons for seasonal variation were also discussed in many studies on this issue. White considers that reduction in physical activity in the winter may be associated with development of DVT.² Boulay et al.¹³ speculate that cold-induced vasoconstriction and reduction in physical activity produce a well documented reduction in blood flow in the lower limbs; in addition, respiratory tract infections may induce hypercoagulability. Other authors believe that change in temperature is an indirect factor and that reduction in day duration and, therefore, sunlight interferes with production of melatonin and coagulability.¹² Wilmshurst¹⁸ adds that plasmatic concentration of fibrinogen is inversely related to room temperature, and part of such increase in fibrinogen in the winter may result from seasonal respiratory infections. In cold conditions, some coagulation factors are increased in vitro such as platelet count and platelet aggregation. Reduced plasma volume and increased blood viscosity during exposure to cold tend to cause thrombosis. However, little is known about the seasonal fluctuation of coagulability.¹³

In this study there were no differences in incidence for VTE cases according to trimesters, which correspond to climatic seasons. However, there was in increased incidence in DVT in warm months. Determination of warm and cold months was based on the mean temperature of all 12 months, which is 20 ºC.¹⁷ Such value was not used in other studies, as they were performed in European countries or in the USA, where temperature variations are higher. They could not be matched with the Brazilian climate to define the same cut-off point for temperature. There were no studies conducted in Brazil or in countries with similar climate that could be used in the comparative analysis.

There was no theory in the literature that could explain increased DVT incidence in warmer months. Absence of correlation between incidence of DVT and PTE is even harder to explain, as a parallelism could be expected between both phenomena. As it is difficult to explain increased VTE in cold months in the northern hemisphere, it might be speculated that the heat causes higher dehydration and maintains people in a more sedentary condition, which in turn might cause increase in DVT.

Little is known about seasonal fluctuation of coagulability, and other variables should be analyzed to discard the possibility of temperature being only a confounding factor in such studies.

Conclusion

VTE is a disease that has no well established relation with climatic variations. Influence of temperature on coagulability still needs to be widely studied.

References

• 1. Arcelus JI, Caprini JA, Monreal M, Suárez C, González-Fajardo J. The management and outcome of acute venous thromboembolism: a prospective registry including 4011 patients. J Vasc Surg. 2003;38:916-22.
• 2. White RH. The epidemiology of venous thromboembolism. Circulation. 2003;107(23 Suppl 1):I4-8.
• 3. Silva MC. Tromboembolismo venoso: epidemiologia e fatores de risco. In: Brito CJ, Duque A, Merlo I, Murilo R, Fonseca VL, editores. Cirurgia vascular. Rio de Janeiro: Revinter; 2002. p. 1123-34.
• 4. Maffei FHA. Epidemiologia da trombose venosa profunda e de suas complicações no Brasil. Cir Vasc Angiol 1998;14:5-8.
• 5. Amary J, Coli Jr DF, Pereira M, Bailone S. Embolismo pulmonar - levantamento em 13500 necrópsias. Arq Hosp Santa Casa S Paulo. 1974;20:143-7.
• 6. Heit JA. Venous thromboembolism epidemiology: implications for prevention and management. Semin Thromb Hemost. 2002;28 Suppl 2:3-13.
• 7. Cooke EA, McNally MA, Mollan RAB. Seasonal variations in fatal pulmonary embolism. Several mechanisms contribute. BMJ. 1995;310:129.
• 8. Anderson FA, Spencer FA. Risk factors for venous thromboembolism. Circulation. 2003;107(23 Suppl 1):I9-16.
• 9. Kearon C. Natural history of venous thromboembolism. Circulation. 2003;107(23 Suppl 1):I22-30.
• 10. Bounameaux H, Hicklin L, Desmarais S. Seasonal variation in deep vein thrombosis. BMJ. 1996;312:284-5.
• 11. Gallerani M, Manfredini R, Ricci L, et al. Sudden death from pulmonary thromboembolism: chronobiological aspects. Eur Heart J. 1992;13:661-5.
• 12. Allan TM, Douglas AS. Seasonal variation in deep vein thrombosis. Fatal pulmonary embolism is increased in both autumn and winter. BMJ. 1996;312:1227.
• 13. Boulay F, Berthier F, Schoukroun G, Raybaut C, Gendreike Y, Blaive B. Seasonal variations in hospital admission for deep vein thrombosis and pulmonary embolism: analysis of discharge data. BMJ. 2001;323:601-2.
• 14. Green J, Edwards C. Seasonal variation in the necropsy incidence of massive pulmonary embolism. J Clin Pathol. 1994;47:58-60.
• 15. Galle C, Wautrecht JC, Motte S, et al. Rôle de la saison dans l"incidence de la thrombose veineuse profonde. J Mal Vasc. 1998;23:99-101.
• 16. Stein PD, Kayali F, Olson RE. Analysis of ocurrence of venous thromboembolic disease in the four seasons. Am J Cardiol. 2004;93:511-3.
• ¹⁷
  The Weather Channel [site na Internet]. The Weather Channel Interactive, Inc. Acess: 01/04. http://br.weather.com/weather/climatology/BRXX0232
  » link
• 18. Wilmshurst P. Temperature and cardiovascular mortality. BMJ. 1994;309:1029-30.

Publication Dates

• Publication in this collection
  13 Mar 2009
• Date of issue
  Mar 2009

History

• Received
  07 Mar 2008
• Accepted
  26 Nov 2008